LungX融合高效网络与注意力机制，提升肺炎检测准确率
LungX: A Hybrid EfficientNet-Vision Transformer Architecture with Multi-Scale Attention for Accurate Pneumonia Detection
- 结合EfficientNet多尺度特征与ViT全局建模能力
- 在2万张胸片上达86.5%准确率，AUC提升6.7个百分点
- 注意力图可解释性强，适合临床辅助诊断场景
肺炎仍是全球主要致死原因之一，及时诊断至关重要。本文提出LungX，一种新型混合架构，融合EfficientNet的多尺度特征、CBAM注意力机制与Vision Transformer的全局上下文建模，以增强肺炎检测性能。在20,000张经筛选的胸部X光片（来自RSNA和CheXpert数据集）上评估，LungX达到86.5%准确率和0.943 AUC，相比EfficientNet-B0基线提升6.7个百分点。可视化分析显示其通过可解释的注意力图实现更优病灶定位。未来工作包括多中心验证及架构优化，目标是实现88%准确率，用于临床部署为人工智能诊断助手。
原文摘要 · Abstract (English)
Pneumonia remains a leading global cause of mortality where timely diagnosis is critical. We introduce LungX, a novel hybrid architecture combining EfficientNet's multi-scale features, CBAM attention mechanisms, and Vision Transformer's global context modeling for enhanced pneumonia detection. Evaluated on 20,000 curated chest X-rays from RSNA and CheXpert, LungX achieves state-of-the-art performance (86.5 percent accuracy, 0.943 AUC), representing a 6.7 percent AUC improvement over EfficientNet-B0 baselines. Visual analysis demonstrates superior lesion localization through interpretable attention maps. Future directions include multi-center validation and architectural optimizations targeting 88 percent accuracy for clinical deployment as an AI diagnostic aid.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。